Exploring holistic face processing of the fusiform and occipital face area with incomplete faces
Poster No:
1073
Submission Type:
Late-Breaking Abstract Submission
Authors:
Xi Li1, Yiying Song1, Fang Tian2
Institutions:
1Faculty of Psychology, Beijing Normal University, Beijing, China, 2Psychological Counseling Center, Shanghai University, Shanghai, China
First Author:
Co-Author(s):
Introduction:
The fusiform face area (FFA) and occipital face area (OFA) are core face-selective regions in the human brain (Isabel Gauthier et al., 2000; Kanwisher et al., 1997), yet their functional division in face processing remains debated. The prevailing view is that the FFA specializes in processing invariant and holistic information, while the OFA focuses on local details (Tsantani et al., 2021). Furthermore, the FFA integrates global configurations for holistic processing, whereas the OFA processes individual facial parts (Kanwisher et al., 1998; Liu et al., 2010; Pitcher et al., 2011). However, previous studies have primarily examined complete faces, leaving how these regions process incomplete faces unknown, which may more directly reflect holistic processing. Here, we investigated the functional division between the FFA and OFA by comparing their activation for complete versus incomplete faces. The incomplete facial stimuli were generated using AlexNet, a deep convolutional neural network (DCNN) with hierarchical processing resembling human ventral stream organization (Yamins et al., 2013).
Methods:
Twenty-two adults underwent structural and task-based fMRI with two experiments: facial fragments and occluded faces. To generate stimuli, we used AlexNet. Each of 28 face images was occluded with square mosaics (3×3 to 63×63 pixels) and presented to AlexNet. Activation differences between original and occluded images were computed, averaged across images, and mapped onto the average face image to create five face activation maps (Fig. 1a-b). Marked facial fragments (F1-F5) and their occluded counterparts (O1-O5) served as stimuli (Fig. 1c). We conducted one-way ANOVAs and paired t-tests to compare the activation values of each brain area for the whole face with those for individual facial fragments or occluded faces. Additionally, we calculated the weighted average activation (Song et al., 2013) for complementary incomplete face pairs (e.g., F1&O1), with weights representing the proportion of non-gray regions in the total image, and compared it with whole-face activation.
·stimuli
Results:
The FFA exhibited significantly greater activation for whole faces versus facial fragments (F = 8.251, p < 0.001; Fig. 2a) and occluded faces (F = 2.544, p = 0.028; Fig. 2b), providing evidence that it responds more strongly to complete faces. Conversely, the OFA showed higher activation for facial fragments containing eyes and eyebrows than whole faces (F4 (t = -3.188, p = 0.002); Fig. 2c), and did not exhibit significantly lower activation for occluded faces compared to whole faces (Fig. 2d). In short, the FFA showed significantly greater activation for whole faces than incomplete faces, while the OFA exhibited strong responses to incomplete faces. Furthermore, in the FFA, activation for whole faces significantly surpassed the weighted average activation for complementary face pairs (F = 2.872, p = 0.015; Fig. 2e), whereas the OFA showed no such difference (F = 1.956, p = 0.084; Fig. 2f), with the sum of activation for facial parts not distinct from that for the whole face. Accordingly, these results suggest that the FFA employs a holistic processing strategy for face recognition, with greater activation for whole faces compared to the weighted average for complementary face pairs. In contrast, the OFA may process faces linearly, with each part of the face contributing useful information.
·activation
Conclusions:
Our findings indicate that the FFA prioritizes global facial information and employs a holistic processing strategy, integrating facial features into a unified entirety. Conversely, the OFA encodes individual facial parts in a linear method, emphasizing local feature-based processing. By combining DCNNs and fMRI, we provide a more direct perspective to examine the functional division between the FFA and OFA in face processing.
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 1
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems
Novel Imaging Acquisition Methods:
BOLD fMRI
Perception, Attention and Motor Behavior:
Perception: Visual 2
Keywords:
ADULTS
Cognition
Cortex
Data analysis
FUNCTIONAL MRI
Vision
Other - Face Recognition; DCNN; FFA; OFA
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Provide references using APA citation style.
Kanwisher, N., McDermott, J., & Chun, M. M. (1997). The fusiform face area: a module in human extrastriate cortex specialized for face perception. The Journal of neuroscience: the official journal of the Society for Neuroscience, 17(11), 4302–4311.
Kanwisher, N., Tong, F., & Nakayama, K. (1998). The effect of face inversion on the human fusiform face area. Cognition, 68(1), B1–B11.
Liu, J., Harris, A., & Kanwisher, N. (2010). Perception of face parts and face configurations: an FMRI study. Journal of cognitive neuroscience, 22(1), 203–211.
Pitcher, D., Walsh, V., & Duchaine, B. (2011). The role of the occipital face area in the cortical face perception network. Experimental brain research, 209(4), 481–493.
Song, Y., Luo, Y. L., Li, X., Xu, M., & Liu, J. (2013). Representation of contextually related multiple objects in the human ventral visual pathway. Journal of cognitive neuroscience, 25(8), 1261–1269.
Tsantani, M., Kriegeskorte, N., Storrs, K., Williams, A. L., McGettigan, C., & Garrido, L. (2021). FFA and OFA Encode Distinct Types of Face Identity Information. The Journal of neuroscience: the official journal of the Society for Neuroscience, 41(9), 1952–1969.
Yamins, D. L., Hong, H., Cadieu, C., & DiCarlo, J. J. (2013). Hierarchical modular optimization of convolutional networks achieves representations similar to macaque IT and human ventral stream. Advances in Neural Information Processing Systems, 26.